3316 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
lower recovery (Filippov et al. 2018). It is possible that
similarly, indiscriminate adsorption of the terpenic ring
onto Al3+and Fe3+ sites is responsible for poor selectivity
in spodumene flotation with high rosin content collectors.
This study compares the selectivity of two commercial
TOFA collectors and two rosin products removed during
the TOFA production process in spodumene flotation. The
TOFA collectors FA1 and FA2 contained 2.4% and 0.8%
rosins, respectively. The two tall oil rosin (TOR) products
NCY and HYR differed in composition with PAN num-
bers (percentage of palustric, abietic, and neoabietic acids)
of 37.3 and 31.6, respectively. The objective of this investi-
gation was to understand if and how the type of rosin acid
impurity within TOFA collectors influences the perfor-
mance of spodumene rougher flotation to guide the design
of more selective fatty acid spodumene collectors.
MATERIALS AND METHODS
Spodumene Ore Sample
An ore sample from a North American spodumene pegma-
tite deposit was used for all batch flotation testwork. The
feed material was first passed through dense media separa-
tion at a crush size of –6.3 mm, and the resulting DMS
middlings and bypass material were combined as the flo-
tation feed. The feed was stage-ground to 100% passing
300 µm with a final P80 of 210 µm and passed through
several bulk upstream processing stages to reject problem-
atic gangue minerals expected to report to the spodumene
concentrate. These stages included wet high-intensity mag-
netic separation (WHIMS) at 10,000 Gauss, desliming to
remove particles 15 µm, and mica pre-flotation with an
amine collector at pH 10.5. The mica flotation tails were
dried, combined, homogenized, and split into 1-kg charges
for spodumene flotation testing.
The chemical analysis of the direct and spodumene flo-
tation feed is presented in Table 1. The spodumene flota-
tion feed contained 1.10% Li2O and 0.75% Fe2O3. The
upstream stages rejected 24.6% of the mass and 69.4% of
the iron, with an overall lithium loss of only 15.9%. Of the
total upstream lithium losses, about 7.6% of the lithium
was lost to the WHIMS magnetic fraction, 5.3% to the
mica concentrate, and 3.0% to the slimes. The flotation
recoveries presented in this manuscript are based on the
feed to spodumene flotation stage and do not include the
upstream performance.
Flotation Reagents
The four primary chemicals used in this study were TOFA
collectors ‘FA1’ and ‘FA2’ and TOR products ‘NCY’ and
‘HYR’. Each TOFA was tested as a pure product and
blended with each TOR individually at levels of 5%. A
total of six collectors/collector blends were evaluated. A
summary of the rosin content, major rosin acid species in
each product, and the tested blends is provided in Table 2.
Because distillation can remove palustric, abietic, and neoa-
bietic acids, FA1 and FA2 have very low PAN numbers of
0.014 and 0.008, respectively, while the blends with the
5% TOR exhibited higher PAN numbers around 1.9 with
5% NCY and around 1.6 with 5% HYR. Of the remaining
rosin species, dehydroabietic acid was highest in both TOR
samples, while secodehydroabietic and isopimaric acids
were highest in FA1 and FA2. In addition to the TOFA
and rosin products at the focus of this study, other flotation
reagents included sodium hydroxide (NaOH) and soda ash
(Na2CO3) as pH modifiers and Pionera F220 (a chemically
modified natural polymer) as a dispersant used during the
alkaline scrubbing stage.
Spodumene Flotation Procedure
A total of seven 1-kg batch flotation tests were performed
in this study. Each test included alkaline scrubbing, dewa-
tering, high-density conditioning, and rougher flotation.
Alkaline scrubbing was performed at pH 11.5 for 10 min-
utes at 50% solids (w/w), with 250 g/t Pionera F220 and
approximately 150 g/t NaOH. The purpose of the scrub-
bing stage was to “activate” the spodumene surface and
mitigate negative effects of surface weathering by break-
ing Al-O bonds and revealing cationic aluminum sites for
collector adsorption on the spodumene surface (Falconer
1949, Moon &Fuerstenau 2003). Dewatering was then
required to reduce the pH from 11.5 down to 8.3 for batch
flotation and to increase the %solids to 60% for high-
density TOFA conditioning. This was achieved by diluting
the scrubbed feed to 11-L with tap water in an acrylic cylin-
der, allowing the particles to settle for 7 minutes, and then
siphoning off the clarified water. The diluting and siphon-
ing process was repeated three times with each sample to
Table 1. ICP and XRF chemical analysis of the direct head sample and spodumene flotation feed
Stream
Assays %
Li Li2O SiO2 Al2O3 K2O Na2O CaO MgO MnO P2O5 Fe2O3
Direct Feed 0.48 0.98 72.1 15.7 1.93 3.85 1.27 0.62 0.10 0.30 1.85
Spodumene Float Feed 0.51 1.10 76.7 14.0 1.29 4.31 0.93 0.20 0.05 0.21 0.75
lower recovery (Filippov et al. 2018). It is possible that
similarly, indiscriminate adsorption of the terpenic ring
onto Al3+and Fe3+ sites is responsible for poor selectivity
in spodumene flotation with high rosin content collectors.
This study compares the selectivity of two commercial
TOFA collectors and two rosin products removed during
the TOFA production process in spodumene flotation. The
TOFA collectors FA1 and FA2 contained 2.4% and 0.8%
rosins, respectively. The two tall oil rosin (TOR) products
NCY and HYR differed in composition with PAN num-
bers (percentage of palustric, abietic, and neoabietic acids)
of 37.3 and 31.6, respectively. The objective of this investi-
gation was to understand if and how the type of rosin acid
impurity within TOFA collectors influences the perfor-
mance of spodumene rougher flotation to guide the design
of more selective fatty acid spodumene collectors.
MATERIALS AND METHODS
Spodumene Ore Sample
An ore sample from a North American spodumene pegma-
tite deposit was used for all batch flotation testwork. The
feed material was first passed through dense media separa-
tion at a crush size of –6.3 mm, and the resulting DMS
middlings and bypass material were combined as the flo-
tation feed. The feed was stage-ground to 100% passing
300 µm with a final P80 of 210 µm and passed through
several bulk upstream processing stages to reject problem-
atic gangue minerals expected to report to the spodumene
concentrate. These stages included wet high-intensity mag-
netic separation (WHIMS) at 10,000 Gauss, desliming to
remove particles 15 µm, and mica pre-flotation with an
amine collector at pH 10.5. The mica flotation tails were
dried, combined, homogenized, and split into 1-kg charges
for spodumene flotation testing.
The chemical analysis of the direct and spodumene flo-
tation feed is presented in Table 1. The spodumene flota-
tion feed contained 1.10% Li2O and 0.75% Fe2O3. The
upstream stages rejected 24.6% of the mass and 69.4% of
the iron, with an overall lithium loss of only 15.9%. Of the
total upstream lithium losses, about 7.6% of the lithium
was lost to the WHIMS magnetic fraction, 5.3% to the
mica concentrate, and 3.0% to the slimes. The flotation
recoveries presented in this manuscript are based on the
feed to spodumene flotation stage and do not include the
upstream performance.
Flotation Reagents
The four primary chemicals used in this study were TOFA
collectors ‘FA1’ and ‘FA2’ and TOR products ‘NCY’ and
‘HYR’. Each TOFA was tested as a pure product and
blended with each TOR individually at levels of 5%. A
total of six collectors/collector blends were evaluated. A
summary of the rosin content, major rosin acid species in
each product, and the tested blends is provided in Table 2.
Because distillation can remove palustric, abietic, and neoa-
bietic acids, FA1 and FA2 have very low PAN numbers of
0.014 and 0.008, respectively, while the blends with the
5% TOR exhibited higher PAN numbers around 1.9 with
5% NCY and around 1.6 with 5% HYR. Of the remaining
rosin species, dehydroabietic acid was highest in both TOR
samples, while secodehydroabietic and isopimaric acids
were highest in FA1 and FA2. In addition to the TOFA
and rosin products at the focus of this study, other flotation
reagents included sodium hydroxide (NaOH) and soda ash
(Na2CO3) as pH modifiers and Pionera F220 (a chemically
modified natural polymer) as a dispersant used during the
alkaline scrubbing stage.
Spodumene Flotation Procedure
A total of seven 1-kg batch flotation tests were performed
in this study. Each test included alkaline scrubbing, dewa-
tering, high-density conditioning, and rougher flotation.
Alkaline scrubbing was performed at pH 11.5 for 10 min-
utes at 50% solids (w/w), with 250 g/t Pionera F220 and
approximately 150 g/t NaOH. The purpose of the scrub-
bing stage was to “activate” the spodumene surface and
mitigate negative effects of surface weathering by break-
ing Al-O bonds and revealing cationic aluminum sites for
collector adsorption on the spodumene surface (Falconer
1949, Moon &Fuerstenau 2003). Dewatering was then
required to reduce the pH from 11.5 down to 8.3 for batch
flotation and to increase the %solids to 60% for high-
density TOFA conditioning. This was achieved by diluting
the scrubbed feed to 11-L with tap water in an acrylic cylin-
der, allowing the particles to settle for 7 minutes, and then
siphoning off the clarified water. The diluting and siphon-
ing process was repeated three times with each sample to
Table 1. ICP and XRF chemical analysis of the direct head sample and spodumene flotation feed
Stream
Assays %
Li Li2O SiO2 Al2O3 K2O Na2O CaO MgO MnO P2O5 Fe2O3
Direct Feed 0.48 0.98 72.1 15.7 1.93 3.85 1.27 0.62 0.10 0.30 1.85
Spodumene Float Feed 0.51 1.10 76.7 14.0 1.29 4.31 0.93 0.20 0.05 0.21 0.75